A prospective study of 3-day versus 2-week immobilization period after anterior cruciate ligament reconstruction

1. Introduction 

Recent innovations in surgical techniques and arthroscopic instruments for ACL reconstruction have enabled aggressive rehabilitation, including early range of motion (ROM) exercise. These concepts have been applied to not only the patellar tendon–bone graft procedure [1] but also to the rehabilitation using the multistranded hamstring tendons procedure. However, we believe that rehabilitation using the multistranded hamstring tendons procedure should be less aggressive generally [2], [3]. ROM exercise was carried out at our hospital from the 3rd postoperative day after ACL reconstruction using multistranded hamstring tendons for the purpose of reducing acute inflammation [4], postoperative pain, and the risk of infection. The potential advantages of an early ROM exercise after ACL reconstruction using multistranded hamstring tendons are numerous, including reducing disuse effects, preventing capsular contractures, maintaining articular cartilage nutrition, and allowing early controlled forces on healing collagen tissues [5]. On the other hand, it is possible that early ROM exercise could increase the amount of graft motion in the bone tunnel [6] and delay the fixation between the grafted tendon and bone for the following reason: The creation of bone tunnels for ACL reconstruction usually leads to the formation of local hematoma at the interface between the bone and the grafted tendon. The hematoma can be kept in place, provide an organized fibrin meshwork, and promote an early inflammatory response with the associated release of mesenchymal cells, cytokines, and growth factors [7]. That would have the positive effect of fixing the grafted tendon and bone. Therefore, we believe that early commencement of ROM exercises could cause the graft to not biologically fixed at the bone tunnels due to early stress on the graft. That might potentially encourage graft motion, creating a “windshield-wiper” or “bungee” effect [8], [9], [10]. Consequently, these effects might induce bone tunnel enlargement or elongation of the graft construction complex and lead to a poor outcome after ACL reconstruction using multistranded hamstring tendons. However, from the current literature, these effects do not appear to correlate with a poor outcome after ACL reconstruction [11]. We hypothesized that the delay in starting ROM exercise would promote the biological fixation of the graft within the bone tunnel and have the positive effective of improving the knee joint stability and position sense, but this suggestion has not been validated. To our knowledge, a comparative study to test this hypothesis has not been published. Therefore, we designed a prospective study to assess the outcome of a 3-day immobilization period after ACL reconstruction using multistranded hamstring tendons compared with that of a 2-week immobilization in this study. Rodeo et al. reported that the interface between the bone and the grafted tendon already comprised histologically of a vascular, highly cellular, fibrous tissue at 2 weeks after transplantation [12]. This was the reason why we chose a 2-week immobilization period for the control group in this study. The purpose of this comparative study was to evaluate whether the delay in starting ROM exercise is indeed associated with improving knee joint stability and position sense.

3. Operative technique 

One of the senior authors (M.O.) performed the surgery arthroscopically using the anterolateral and anteromedial portals. A 4-cm skin incision was made on the medial side of the proximal tibia, over the insertion of the pes anserinus. For single-bundle reconstruction, we quadrupled the semitendinosus tendon. The proximal ends were fixed using an Endobutton CL (Acufex; Smith and Nephew, Mansfield, Massachusetts), and the distal ends were sutured with Endobutton tape (Acufex; Smith and Nephew). The position of the femoral tunnel was 5 mm anterior to the posterior margin of the intercondylar notch and at the 10 o'clock (right) or 2 o'clock (left) position with respect to the apex of the notch. The Endobutton was fixed to the lateral cortex of the femur. We applied a tension force of 50 N to the distal Endobutton tape of the graft and secured it with the double-stapling technique with the knee at 90° of flexion.

4. Postoperative rehabilitation program 

Active quadriceps exercises were carried out as soon as possible. In both groups, the knee was supported for 2 days in a knee brace, allowing 20° of flexion. 1) 3D group, which started ROM exercise on the 3rd postoperative day. 2) 2W group, which was immobilized continuously for a 2-week period and started ROM exercise at the end of the first 2 postoperative weeks. ROM exercise was done using a continuous passive motion device. A soft knee brace prevented the knee from attaining full extension for 3 months. Full extension was allowed after removal of the soft knee brace. All post-operative rehabilitation programs were equal in both groups after the first 2 postoperative weeks. Partial weight bearing was allowed at 2 weeks, followed by full weight bearing at 4 weeks. After that, patients were allowed to leave the hospital. Jogging was recommended at 3 months.

5. Evaluation 

The clinical outcome was graded according to the Lysholm knee scoring scale [13]. We examined the knees and determined the Lysholm scores preoperatively and postoperatively. Measurements of stability and proprioception [14], [15] were carried out before operation and at 3, 6, and 12 months after surgery. Anterior displacement of the tibia was measured using the KT-2000 knee arthrometer (MEDmetric, San Diego, California) with the knee at 20° of flexion. An anterior displacement force of 133 N was applied to the tibia of both the injured and normal knees, and the difference was recorded in millimeters. Tests of position sense were performed according to the method of Skinner, Barrack and Cook using a Biodex system (Biodex Co. New York, NY). Inaccuracy was recorded as the difference between the perceived angle and the actual angle of flexion. The test was carried out 10 times in each knee at low angles (5° and 25°). The mean inaccuracy at low angles was calculated for each knee. The final inaccuracy was expressed as the difference between the mean scores for the injured and the normal knee.

For analysis of thigh muscle strength, both extension and flexion peak torques were isokinetically measured preoperatively and at 6, and 12 months after surgery using the Biodex system (Biodex Co, New York, NY) at 60°/s. The patients were instructed and assisted by a physical therapist so they would obtain their maximum strength. Each isokinetic measurement consisted of five cycles of continuous extension–flexion motion at maximal strength, and the best torque of the five repetitions was used for analysis. The value of the side-to-side ratio in the peak torque for each group was used as the representative parameter for thigh muscle strength.

6. Statistical analysis 

The Wilcoxon signed rank test was used to compare the pre- and postoperative data of the KT-2000 knee arthrometer of the group. Repeated analysis of variance (ANOVA) measures were used for intergroup comparisons of longitudinal data. A p value of less than 0.05 was regarded as being statistically significant. All statistical analyses were carried out on Statview 5.0 (SAS Institite, Cary, North Carolina).

7. Results 

7.2. Knee joint stability 

The mean side-to-side differences of anterior laxity with the knee at 20° of flexion in the 3D and 2W groups were 5.4±2.3 mm and 5.8±2.7 mm before surgery. In the 3D group, the mean side-to-side differences of anterior knee displacement were 0.3±1.6 mm at 3 months, 0.6±2.5 mm at 6 months, and 0.9±2.7 mm at 12 months after surgery. In contrast, the mean side-to-side differences in the 2W group were −0.5±1.3 mm at 3 months, 0.3±2.4 mm at 6 months, and 0.7±1.9 mm at 12 months after surgery. An average of 5.4±2.3 mm preoperatively significantly improved to an average of 0.9±2.7 mm in the 3D group, while an average of 5.8±2.7 mm also significantly improved to 0.7±1.9 mm in the 2W group. There was no statistically significant difference in outcome between the two groups at any time point after surgery (Fig. 1).

View Large Image Download to PowerPoint

Fig. 1. The side-to-side differences of anterior laxity with the knee at 20° of flexion in each group. ⁎ There were no statistically significant differences in outcomes between the two groups at any time point after surgery.

7.3. Joint position sense 

The final inaccuracy in the 3D and 2W groups, respectively, was 1.9.±1.5° and 1.7±1.9° before operation and 1.2±1.4° and 1.1±1.0° at the final follow-up. There was no statistically significant difference between the two groups at any time point after surgery (Fig. 2).

View Large Image Download to PowerPoint

Fig. 2. The final inaccuracy at low angles in each group. ⁎ There were no statistically significant differences between the two groups at any time point after surgery.

7.4. Thigh muscle strength 

For the isokinetic measurements of knee extension, the ratios of the peak torque relative to the control side in the 3D and 2W groups were 0.65±9.5 and 0.62±7.4 at 6 months after surgery and 0.83±11.2 and 0.81±10.6 at 12 months after surgery. For the isokinetic measurements of knee flexion, the ratios were 0.69±8.4 and 0.66±7.8 at 6 months after surgery and 0.88±10.3 and 0.86±11.2 at 12 months after surgery. There were no statistically significant differences between the two groups in isokinetic knee extension and flexion torques at any time point after surgery (Fig. 3, Fig. 4).

View Large Image Download to PowerPoint

Fig. 3. The isokinetic muscle strength in knee extension: side-to-side difference. ⁎ There were no statistically significant differences between the two groups in isokinetic knee extension.

View Large Image Download to PowerPoint

Fig. 4. The isokinetic muscle strength in knee flexion: side-to-side difference. ⁎ There were no statistically significant differences between the two groups in isokinetic knee flexion.

8. Discussion 

Development of arthroscopic instruments and devices over the past 20 years has enabled surgeons to provide good stable knees that allow their patients to participate in postoperative aggressive early rehabilitation, with minimal immobilization of the knee, and early active mobilization. Paulos et al. reported the results of their postoperative rehabilitation program after ACL reconstruction with a bone–patellar tendon–bone graft [16]. They used an above-the-knee cast with the knee flexed 30° to 60° for 8 weeks. Shelbourne and Gray described an accelerated rehabilitation protocol after ACL reconstruction using a bone–patellar tendon–bone graft which allowed full extension from the 1st postoperative day [17]. They concluded that the patients who followed their rehabilitation program obtained long-term stability and were able to return to full sports activities. Yasuda and Sasaki described the postoperative rehabilitation protocol after ACL reconstruction using multistranded hamstring tendons [18]. Their protocol was based on the results of their previous biomechanical studies [19]. ROM exercise (0–90°) was allowed immediately after surgery in their program. Although they reported good clinical results, these studies were retrospective.

In this study, we prospectively compared the clinical results of 3-day immobilization after surgery, with those of 2-week immobilization. However, our study has shown no significant differences in the clinical outcome between the two groups in terms of the subjective knee function, joint stability, position sense, and thigh muscle strength. In our results, the mean side-to-side differences of anterior laxity indicated that there was an initial improvement 3 months after surgery in terms of improvement in stability in both groups with a subsequent less of an improvement over time. It could be due to the stretching out of the ACL graft over time. However, the mean side-to-side differences of anterior laxity would become stable after 24 months' follow-up as well as previous reports [8], [20]. Fujimoto et al. [21] reported the first serial arthrometer measurements of ACL reconstruction using hamstring tendons. At the 24-month follow-up evaluation, 37 patients (74%) had a functional reconstruction and normal displacement measurements. The percentage of grafts that stretched but did not fail was low (12/50, 24%). These knees, with 3 to 5 mm of increased displacement, had positive glide pivot shift tests but were still considered partially functional. A low failure rate (1/50 patients, 2%) was found after surgery. Our results were favorable when compared with those of others [22], [23] who provided arthrometer measurements after ACL reconstruction using hamstring tendons. To our knowledge, there have been two reports that have described prospective randomized comparisons of the different postoperative immobilization periods after ACL reconstruction using a bone–patellar tendon–bone graft. Noyes et al. prospectively compared the clinical results of 2-day immobilization and 7-day immobilization after ACL reconstruction using a bone–patellar tendon–bone graft [24]. They found no significant differences between the two groups, except that the loss of thigh circumference was more notable after the 7-day immobilization and so they concluded that the 2-day immobilization was better. Henriksson et al. reported the clinical outcome of a study which compared the 7-day immobilization and 5-week immobilization after ACL reconstruction using bone–patellar tendon–bone grafts [25]. Their results showed that there were no significant differences between the groups in terms of the knee joint stability, the subjective knee function, or activity level. The conclusion is that the results of the early ROM exercise group are as good as those for the 5-week immobilization group. However, the muscle strength deficit in the hamstring muscles was significantly larger in the early ROM exercise group.

There have been no reports which have compared the different postoperative immobilization periods after ACL reconstruction using multistranded hamstring tendons prospectively. Although the joint stability and position sense of the 2W group were predisposed to be superior to that of the 3D group, our study has shown no significant differences statistically in the clinical outcomes between the two groups in terms of the subjective knee function, joint stability, position sense, and thigh muscle strength. There is a concern with our study. The number of patients included was relatively small. If the sample size were larger, a difference may have been detected between the groups. In our study, we could not find any advantages of the 2-week immobilization period in terms of joint stability, joint position sense, and thigh muscle strength over the 3-day immobilization period in our rehabilitation program. That is due to the fact that we have never carried out our rehabilitation program aggressively. As Henriksson et al. have reported, it is suggested that early ROM exercise should be accompanied by a careful rehabilitation program, which confirms the recovery of joint stability, and thigh muscle strength [26]. Indeed, it has been carried out carefully since the introduction of ROM exercise, with measurements of joint stability, joint position sense, and thigh muscle strength in our rehabilitation program [27]. Therefore, we could achieve good clinical results in both groups. We believe that immediately after surgery is the critical period for biological healing between the grafted tendon and bone because the creation of bone tunnels for ACL reconstruction probably leads to associated release of mesenchymal cells, cytokines, and growth factors.

Therefore, we concluded that the post-operative 3-day immobilization period is preferable to the longer 2-week period in our rehabilitation program from the viewpoint of the psychological load on the patients and the duration of hospitalization after ACL reconstruction using multistranded hamstring tendons.